Improving ground truth measurement using microscopic simulator for urban network

Abstract

Transportation management center (TMC) operations need to enhance the ability of operators to mange traffic in a more responsive and effective way from real-time traffic flow data. The traffic data collected on an actual road network are sent to the TMC for collection and then provided to various information provision media (Traffic information board, VMS, ARS, etc.) through primary and secondary information processing. However, accuracy for the traffic information extracted from such systems has not been clearly known, and one of the main reasons results from reliability for ground truth (GT). In order to verify accuracy and reliability for the traffic information processed in the TMC, microscopic traffic flow simulation will be used as GT, unlike the general methods. This paper proposes using microscopic traffic flow simulation as an alternative for improving the limitations in the ground truth survey (GTS). The main objective of this paper is to examine a feasibility of microscopic simulation in order to improve the reliability for the GTS. The microscopic traffic flow simulation can extract vehicle detection data through simulation considering various traffic environments. In particular, the simulation is advantageous in that it is possible to extract spot traffic data, space traffic data, vehicle trajectory data and the like by preparing various scenarios considering detection system types, traffic, geometry, signal operation, etc. Therefore, it is decided that the microscopic simulation would be highly usable as an alternative to the GTS in terms of efficiency and reliability if it simulates the traffic situation in the real world as similar as possible.